Study on landslide susceptibility mapping based on rock-soil characteristic factors

Area of Science:

• Geosciences
• Environmental Science
• Geotechnical Engineering

Background:

• Landslide Susceptibility Mapping (LSM) is crucial for hazard mitigation in geologically active regions.
• Traditional LSM methods often rely on basic geological factors, potentially overlooking critical rock-soil characteristics.
• The Three Gorges Reservoir Area faces significant landslide risks, necessitating advanced predictive models.

Purpose of the Study:

• To introduce and evaluate four novel rock-soil characteristic factors for improved Landslide Susceptibility Mapping (LSM).
• To compare the predictive performance of models incorporating these new factors against traditional methods.
• To assess the influence of intrinsic versus external rock-soil attributes on LSM accuracy.

Main Methods:

• Development of four rock-soil characteristic factors: Lithology, Rock Structure, Rock Infiltration, and Rock Weathering.
• Integration of these factors with 11 basic geological factors to create diverse input datasets for LSM.
• Application of Logistic Regression, Artificial Neural Network, and Support Vector Machine for LSM modeling.
• Quantitative evaluation using Specific Category Precision, Receiver Operating Characteristic (ROC) curves, and other statistical metrics.

Main Results:

• Models incorporating Rock Structure factors demonstrated superior performance compared to those based on Lithology.
• Intrinsic attribute factors (Rock Structure, Rock Infiltration, Rock Weathering) showed higher significance in improving LSM accuracy than external participation factors.
• The proposed methodology effectively enhanced the scientificity, accuracy, and validity of Landslide Susceptibility Mapping.

Conclusions:

• Rock-soil characteristic factors, particularly Rock Structure and intrinsic attributes, are vital for accurate LSM.
• The novel approach offers a more scientifically robust and valid method for predicting landslide susceptibility.
• This research provides valuable insights for landslide risk management in the Three Gorges Reservoir Area and similar regions.